Towards an ontology-based dictionary for production planning and control in the domain of injection molding as a basis for standardized asset administration shells

Abstract

The use of digital technologies in the industrial environment enables great potential for increasing efficiency in manufacturing. One building block are production environments that plan and control their production flow autonomously and decentrally. To this end, all machines and systems (so-called “assets”) need to communicate with each other and derive suitable actions based on the exchanged information. Therefore, all assets need to be represented in the virtual world. This can be realized with digital twins. A concrete implementation of digital twins is the asset administration shell, which comprises all the assets’ properties and the endpoint of the corresponding asset, so intercommunication is possible. Here, the challenges comprise establishing a manufacturer-independent vocabulary that standardizes the assets’ properties and enabling the machines and systems to interpret this vocabulary semantically. Existing standards and information models represent only a fraction of the information requirements (i.e., terms) in this domain, making autonomous production planning and control (PPC) challenging to implement. Furthermore, the information requirements of the machines and peripheral assets as well as the corresponding information flows are insufficiently defined. Therefore, this contribution aims to build a comprehensive vocabulary for the domain of PPC, which serves as a basis for standardized asset administration shells that realize machine-to-machine communication. In particular, PPC processes concerning the injection molding domain’s characteristics are considered since the interaction between the domain’s assets, i.e., injection molding machines, molds, peripheral assets, raw materials, and operators, are especially complex. For this purpose, the relevant input and output information within the injection molding domain was first collected for each process step in PPC. After that, a UML class diagram was modeled under consideration of established standards. The result of this work is an ontology, which can be used as a dictionary for the PPC in the injection molding domain and as a foundation of standardized digital twins in the form of asset administration shells.